A compute-in-memory chip based on resistive random-access memory.

Realizing increasingly complex artificial intelligence (AI) functionalities directly on edge devices calls for unprecedented energy efficiency of edge hardware. Compute-in-memory (CIM) based on resistive random-access memory (RRAM)1 promises to meet such demand by storing AI model weights in dense, analogue and non-volatile RRAM devices, and by performing AI computation directly within RRAM, thus eliminating power-hungry data movement between separate compute and memory2-5. Although recent studies have demonstrated in-memory matrix-vector multiplication on fully integrated RRAM-CIM hardware6-17, it remains a goal for a RRAM-CIM chip to simultaneously deliver high energy efficiency, versatility to support diverse models and software-comparable accuracy. Although efficiency, versatility and accuracy are all indispensable for broad adoption of the technology, the inter-related trade-offs among them cannot be addressed by isolated improvements on any single abstraction level of the design. Here, by co-optimizing across all hierarchies of the design from algorithms and architecture to circuits and devices, we present NeuRRAM-a RRAM-based CIM chip that simultaneously delivers versatility in reconfiguring CIM cores for diverse model architectures, energy efficiency that is two-times better than previous state-of-the-art RRAM-CIM chips across various computational bit-precisions, and inference accuracy comparable to software models quantized to four-bit weights across various AI tasks, including accuracy of 99.0 percent on MNIST18 and 85.7 percent on CIFAR-1019 image classification, 84.7-percent accuracy on Google speech command recognition20, and a 70-percent reduction in image-reconstruction error on a Bayesian image-recovery task.

Wafer-scale single-crystal hexagonal boron nitride monolayers on Cu (111).

Ultrathin two-dimensional (2D) semiconducting layered materials offer great potential for extending Moore's law of the number of transistors in an integrated circuit1. One key challenge with 2D semiconductors is to avoid the formation of charge scattering and trap sites from adjacent dielectrics. An insulating van der Waals layer of hexagonal boron nitride (hBN) provides an excellent interface dielectric, efficiently reducing charge scattering2,3. Recent studies have shown the growth of single-crystal hBN films on molten gold surfaces4 or bulk copper foils5. However, the use of molten gold is not favoured by industry, owing to its high cost, cross-contamination and potential issues of process control and scalability. Copper foils might be suitable for roll-to-roll processes, but are unlikely to be compatible with advanced microelectronic fabrication on wafers. Thus, a reliable way of growing single-crystal hBN films directly on wafers would contribute to the broad adoption of 2D layered materials in industry. Previous attempts to grow hBN monolayers on Cu (111) metals have failed to achieve mono-orientation, resulting in unwanted grain boundaries when the layers merge into films6,7. Growing single-crystal hBN on such high-symmetry surface planes as Cu (111)5,8 is widely believed to be impossible, even in theory. Nonetheless, here we report the successful epitaxial growth of single-crystal hBN monolayers on a Cu (111) thin film across a two-inch c-plane sapphire wafer. This surprising result is corroborated by our first-principles calculations, suggesting that the epitaxial growth is enhanced by lateral docking of hBN to Cu (111) steps, ensuring the mono-orientation of hBN monolayers. The obtained single-crystal hBN, incorporated as an interface layer between molybdenum disulfide and hafnium dioxide in a bottom-gate configuration, enhanced the electrical performance of transistors. This reliable approach to producing wafer-scale single-crystal hBN paves the way to future 2D electronics.

Graphene and two-dimensional materials for silicon technology.

The development of silicon semiconductor technology has produced breakthroughs in electronics-from the microprocessor in the late 1960s to early 1970s, to automation, computers and smartphones-by downscaling the physical size of devices and wires to the nanometre regime. Now, graphene and related two-dimensional (2D) materials offer prospects of unprecedented advances in device performance at the atomic limit, and a synergistic combination of 2D materials with silicon chips promises a heterogeneous platform to deliver massively enhanced potential based on silicon technology. Integration is achieved via three-dimensional monolithic construction of multifunctional high-rise 2D silicon chips, enabling enhanced performance by exploiting the vertical direction and the functional diversification of the silicon platform for applications in opto-electronics and sensing. Here we review the opportunities, progress and challenges of integrating atomically thin materials with silicon-based nanosystems, and also consider the prospects for computational and non-computational applications.

Probing the Melting Transitions in Phase-Change Superlattices via Thin Film Nanocalorimetry.

Phase-change superlattices with nanometer thin sublayers are promising for low-power phase-change memory (PCM) on rigid and flexible platforms. However, the thermodynamics of the phase transition in such nanoscale superlattices remain unexplored, especially at ultrafast scanning rates, which is crucial for our fundamental understanding of superlattice-based PCM. Here, we probe the phase transition of Sb2Te3 (ST)/Ge2Sb2Te5 (GST) superlattices using nanocalorimetry with a monolayer sensitivity (∼1 Å) and a fast scanning rate (105 K/s). For a 2/1.8 nm/nm Sb2Te3/GST superlattice, we observe an endothermic melting transition with an ∼240 °C decrease in temperature and an ∼8-fold decrease in enthalpy compared to those for the melting of GST, providing key thermodynamic insights into the low-power switching of superlattice-based PCM. Nanocalorimetry measurements for Sb2Te3 alone demonstrate an intrinsic premelting similar to the unique phase transition of superlattices, thus revealing a critical role of the Sb2Te3 sublayer within our superlattices. These results advance our understanding of superlattices for energy-efficient data storage and computing.

Three-dimensional integration of nanotechnologies for computing and data storage on a single chip.

The computing demands of future data-intensive applications will greatly exceed the capabilities of current electronics, and are unlikely to be met by isolated improvements in transistors, data storage technologies or integrated circuit architectures alone. Instead, transformative nanosystems, which use new nanotechnologies to simultaneously realize improved devices and new integrated circuit architectures, are required. Here we present a prototype of such a transformative nanosystem. It consists of more than one million resistive random-access memory cells and more than two million carbon-nanotube field-effect transistors-promising new nanotechnologies for use in energy-efficient digital logic circuits and for dense data storage-fabricated on vertically stacked layers in a single chip. Unlike conventional integrated circuit architectures, the layered fabrication realizes a three-dimensional integrated circuit architecture with fine-grained and dense vertical connectivity between layers of computing, data storage, and input and output (in this instance, sensing). As a result, our nanosystem can capture massive amounts of data every second, store it directly on-chip, perform in situ processing of the captured data, and produce 'highly processed' information. As a working prototype, our nanosystem senses and classifies ambient gases. Furthermore, because the layers are fabricated on top of silicon logic circuitry, our nanosystem is compatible with existing infrastructure for silicon-based technologies. Such complex nano-electronic systems will be essential for future high-performance and highly energy-efficient electronic systems.

Unveiling the Effect of Superlattice Interfaces and Intermixing on Phase Change Memory Performance.

Superlattice (SL) phase change materials have shown promise to reduce the switching current and resistance drift of phase change memory (PCM). However, the effects of internal SL interfaces and intermixing on PCM performance remain unexplored, although these are essential to understand and ensure reliable memory operation. Here, using nanometer-thin layers of Ge2Sb2Te5 and Sb2Te3 in SL-PCM, we uncover that both switching current density (Jreset) and resistance drift coefficient (v) decrease as the SL period thickness is reduced (i.e., higher interface density); however, interface intermixing within the SL increases both. The signatures of distinct versus intermixed interfaces also show up in transmission electron microscopy, X-ray diffraction, and thermal conductivity measurements of our SL films. Combining the lessons learned, we simultaneously achieve low Jreset ≈ 3-4 MA/cm2 and ultralow v ≈ 0.002 in mushroom-cell SL-PCM with ∼110 nm bottom contact diameter, thus advancing SL-PCM technology for high-density storage and neuromorphic applications.

Gate Quantum Capacitance Effects in Nanoscale Transistors.

As the physical dimensions of a transistor gate continue to shrink to a few atoms, performance can be increasingly determined by the limited electronic density of states (DOS) in the gate and the gate quantum capacitance (CQ). We demonstrate the impact of gate CQ and the dimensionality of the gate electrode on the performance of nanoscale transistors through analytical electrostatics modeling. For low-dimensional gates, the gate charge can limit the channel charge, and the transfer characteristics of the device become dependent on the gate DOS. We experimentally observe for the first time, room-temperature gate quantization features in the transfer characteristics of single-walled carbon nanotube (CNT)-gated ultrathin silicon-on-insulator (SOI) channel transistors; features which can be attributed to the Van Hove singularities in the one-dimensional DOS of the CNT gate. In addition to being an important aspect of future transistor design, potential applications of this phenomenon include multilevel transistors with suitable transfer characteristics obtained via engineered gate DOS.

Low-Temperature Side Contact to Carbon Nanotube Transistors: Resistance Distributions Down to 10 nm Contact Length.

Carbon nanotube field-effect transistors (CNFETs) promise to improve the energy efficiency, speed, and transistor density of very large scale integration circuits owing to the intrinsic thin channel body and excellent charge transport properties of carbon nanotubes. Low-temperature fabrication (e.g., <400 °C) is a key enabler for the monolithic three-dimensional (3D) integration of CNFET digital logic into a device technology platform that overcomes memory bandwidth bottlenecks for data-abundant applications such as big-data analytics and machine learning. However, high contact resistance for short CNFET contacts has been a major roadblock to establishing CNFETs as a viable technology because the contact resistance, in series with the channel resistance, reduces the on-state current of CNFETs. Additionally, the variation in contact resistance remains unstudied for short contacts and will further degrade the energy efficiency and speed of CNFET circuits. In this work, we investigate by experiments the contact resistance and statistical variation of room-temperature fabricated CNFET contacts down to 10 nm contact lengths. These CNFET contacts are ∼15 nm shorter than the state-of-the-art Si CMOS "7 nm node" contact length, allowing for multiple generations of future scaling of the transistor-contacted gate pitch. For the 10 nm contacts, we report contact resistance values down to 6.5 kΩ per source/drain contact for a single carbon nanotube (CNT) with a median contact resistance of 18.2 kΩ. The 10 nm contacts reduce the CNFET current by as little as 13% at VDS = 0.7 V compared with the best reported 200 nm contacts to date, corroborated by results in this work. Our analysis of RC from 232 single-CNT CNFETs between the long-contact (e.g., 200 nm) and short-contact (e.g., 10 nm) regimes quantifies the resistance variation and projects the impact on CNFET current variability versus the number of CNT in the transistor. The resistance distribution reveals contact-length-dependent RC variations become significant below 20 nm contact length. However, a larger source of CNFET resistance variation is apparent at all contact lengths used in this work. To further investigate the origins of this contact-length-independent resistance variation, we analyze the variation of RC in arrays of identical CNFETs along a single CNT of constant diameter and observe the random occurrence of high  RC, even on correlated CNFETs.

Spatial Separation of Carrier Spin by the Valley Hall Effect in Monolayer WSe2 Transistors.

We investigate the valley Hall effect (VHE) in monolayer WSe2 field-effect transistors using optical Kerr rotation measurements at 20 K. While studies of the VHE have so far focused on n -doped MoS2, we observe the VHE in WSe2 in both the n - and p -doping regimes. Hole doping enables access to the large spin-splitting of the valence band of this material. The Kerr rotation measurements probe the spatial distribution of the valley carrier imbalance induced by the VHE. Under current flow, we observe distinct spin-valley polarization along the edges of the transistor channel. From analysis of the magnitude of the Kerr rotation, we infer a spin-valley density of 44 spins/µm, integrated over the edge region in the p -doped regime. Assuming a spin diffusion length less than 0.1 µm, this corresponds to a spin-valley polarization of the holes exceeding 1%.

Fast Spiking of a Mott VO2-Carbon Nanotube Composite Device.

The recent surge of interest in brain-inspired computing and power-efficient electronics has dramatically bolstered development of computation and communication using neuron-like spiking signals. Devices that can produce rapid and energy-efficient spiking could significantly advance these applications. Here we demonstrate direct current or voltage-driven periodic spiking with sub-20 ns pulse widths from a single device composed of a thin VO2 film with a metallic carbon nanotube as a nanoscale heater, without using an external capacitor. Compared with VO2-only devices, adding the nanotube heater dramatically decreases the transient duration and pulse energy, and increases the spiking frequency, by up to 3 orders of magnitude. This is caused by heating and cooling of the VO2 across its insulator-metal transition being localized to a nanoscale conduction channel in an otherwise bulk medium. This result provides an important component of energy-efficient neuromorphic computing systems and a lithography-free technique for energy-scaling of electronic devices that operate via bulk mechanisms.

Carbon nanotube computer.

The miniaturization of electronic devices has been the principal driving force behind the semiconductor industry, and has brought about major improvements in computational power and energy efficiency. Although advances with silicon-based electronics continue to be made, alternative technologies are being explored. Digital circuits based on transistors fabricated from carbon nanotubes (CNTs) have the potential to outperform silicon by improving the energy-delay product, a metric of energy efficiency, by more than an order of magnitude. Hence, CNTs are an exciting complement to existing semiconductor technologies. Owing to substantial fundamental imperfections inherent in CNTs, however, only very basic circuit blocks have been demonstrated. Here we show how these imperfections can be overcome, and demonstrate the first computer built entirely using CNT-based transistors. The CNT computer runs an operating system that is capable of multitasking: as a demonstration, we perform counting and integer-sorting simultaneously. In addition, we implement 20 different instructions from the commercial MIPS instruction set to demonstrate the generality of our CNT computer. This experimental demonstration is the most complex carbon-based electronic system yet realized. It is a considerable advance because CNTs are prominent among a variety of emerging technologies that are being considered for the next generation of highly energy-efficient electronic systems.

Unipolar n-Type Black Phosphorus Transistors with Low Work Function Contacts.

Black phosphorus (BP) is a promising two-dimensional (2D) material for nanoscale transistors, due to its expected higher mobility than other 2D semiconductors. While most studies have reported ambipolar BP with a stronger p-type transport, it is important to fabricate both unipolar p- and n-type transistors for low-power digital circuits. Here, we report unipolar n-type BP transistors with low work function Sc and Er contacts, demonstrating a record high n-type current of 200 µA/µm in 6.5 nm thick BP. Intriguingly, the electrical transport of the as-fabricated, capped devices changes from ambipolar to n-type unipolar behavior after a month at room temperature. Transmission electron microscopy analysis of the contact cross-section reveals an intermixing layer consisting of partly oxidized metal at the interface. This intermixing layer results in a low n-type Schottky barrier between Sc and BP, leading to the unipolar behavior of the BP transistor. This unipolar transport with a suppressed p-type current is favorable for digital logic circuits to ensure a lower off-power consumption.

In situ visualization and detection of surface potential variation of mono and multilayer MoS2 under different humidities using Kelvin probe force microscopy.

The surface potential (SP) variations in mono and multilayer molybdenum disulfide (MoS2) are visualized in situ and detected using Kelvin probe force microscopy (KPFM) in different humidity conditions for the first time. N-type doping, which originates from the SiO2 substrate, is discovered in the exfoliated MoS2 and is accompanied by a screening length of five layers. The influence of water, which serves as an environmental gating for MoS2, is investigated by controlling the relative humidities (RHs) in the environmental chamber. A monotonic decrease in the SP is observed when the threshold concentration is achieved. This corresponds to the Fermi level variation, which is dominated by different processes. The results also indicate that water adsorption could result in MoS2 p-type doping and provide compensation that partially counteracts the substrate effect. Under this condition, the interlayer screening effect is influenced because of the water dipole-induced electric field. Density functional theory calculations are performed to determine the band structure variations and the interactions between water molecules and between water molecules and the MoS2 surface in mono and trilayer MoS2 under different RHs. The calculations are in excellent agreement with the experimental results. We propose that in situ measurements of the SP using KPFM under different environmental regimes is a noninvasive and effective method to provide real-time visualization and detection of electronic property variations in two-dimensional materials.

Statistical Study on the Schottky Barrier Reduction of Tunneling Contacts to CVD Synthesized MoS2.

Creating high-quality, low-resistance contacts is essential for the development of electronic applications using two-dimensional (2D) layered materials. Many previously reported methods for lowering the contact resistance rely on volatile chemistry that either oxidize or degrade in ambient air. Nearly all reported efforts have been conducted on only a few devices with mechanically exfoliated flakes which is not amenable to large scale manufacturing. In this work, Schottky barrier heights of metal-MoS2 contacts to devices fabricated from CVD synthesized MoS2 films were reduced by inserting a thin tunneling Ta2O5 layer between MoS2 and metal contacts. Schottky barrier height reductions directly correlate with exponential reductions in contact resistance. Over two hundred devices were tested and contact resistances extracted for large scale statistical analysis. As compared to metal-MoS2 Schottky contacts without an insulator layer, the specific contact resistivity has been lowered by up to 3 orders of magnitude and current values increased by 2 orders of magnitude over large area (>4 cm(2)) films.

Removable and Recyclable Conjugated Polymers for Highly Selective and High-Yield Dispersion and Release of Low-Cost Carbon Nanotubes.

High-purity semiconducting single-walled carbon nanotubes (s-SWNTs) with little contamination are desired for high-performance electronic devices. Although conjugated polymer wrapping has been demonstrated as a powerful and scalable strategy for enriching s-SWNTs, this approach suffers from significant contaminations by polymer residues and high cost of conjugated polymers. Here, we present a simple but general approach using removable and recoverable conjugated polymers for separating s-SWNTs with little polymer contamination. A conjugated polymer with imine linkages was synthesized to demonstrate this concept. Moreover, the SWNTs used are without prepurifications and very low cost. The polymer exhibits strong dispersion for large-diameter s-SWNTs with high yield (23.7%) and high selectivity (99.7%). After s-SWNT separation, the polymer can be depolymerized into monomers and be cleanly removed under mild acidic conditions, yielding polymer-free s-SWNTs. The monomers can be almost quantitatively recovered to resynthesize polymer. This approach enables isolation of "clean" s-SWNTs and, at the same time, greatly lowers costs for SWNT separation.

Picosecond Electric-Field-Induced Threshold Switching in Phase-Change Materials.

Many chalcogenide glasses undergo a breakdown in electronic resistance above a critical field strength. Known as threshold switching, this mechanism enables field-induced crystallization in emerging phase-change memory. Purely electronic as well as crystal nucleation assisted models have been employed to explain the electronic breakdown. Here, picosecond electric pulses are used to excite amorphous Ag_{4}In_{3}Sb_{67}Te_{26}. Field-dependent reversible changes in conductivity and pulse-driven crystallization are observed. The present results show that threshold switching can take place within the electric pulse on subpicosecond time scales-faster than crystals can nucleate. This supports purely electronic models of threshold switching and reveals potential applications as an ultrafast electronic switch.

Disturbance characteristics of half-selected cells in a cross-point resistive switching memory array.

Disturbance characteristics of cross-point resistive random access memory (RRAM) arrays are comprehensively studied in this paper. An analytical model is developed to quantify the number of pulses (#Pulse) the cell can bear before disturbance occurs under various sub-switching voltage stresses based on physical understanding. An evaluation methodology is proposed to assess the disturb behavior of half-selected (HS) cells in cross-point RRAM arrays by combining the analytical model and SPICE simulation. The characteristics of cross-point RRAM arrays such as energy consumption, reliable operating cycles and total error bits are evaluated by the methodology. A possible solution to mitigate disturbance is proposed.

A general design strategy for block copolymer directed self-assembly patterning of integrated circuits contact holes using an alphabet approach.

Directed self-assembly (DSA) is a promising lithography candidate for technology nodes beyond 14 nm. Researchers have shown contact hole patterning for random logic circuits using DSA with small physical templates. This paper introduces an alphabet approach that uses a minimal set of small physical templates to pattern all contacts configurations on integrated circuits. We illustrate, through experiments, a general and scalable template design strategy that links the DSA material properties to the technology node requirements.

Energy-Efficient Phase-Change Memory with Graphene as a Thermal Barrier.

Phase-change memory (PCM) is an important class of data storage, yet lowering the programming current of individual devices is known to be a significant challenge. Here we improve the energy-efficiency of PCM by placing a graphene layer at the interface between the phase-change material, Ge2Sb2Te5 (GST), and the bottom electrode (W) heater. Graphene-PCM (G-PCM) devices have ∼40% lower RESET current compared to control devices without the graphene. This is attributed to the graphene as an added interfacial thermal resistance which helps confine the generated heat inside the active PCM volume. The G-PCM achieves programming up to 10(5) cycles, and the graphene could further enhance the PCM endurance by limiting atomic migration or material segregation at the bottom electrode interface.